Evaluation of the accuracy of the average Mu-values within patients from MR derived Mu-maps

Abstract

We describe an evaluation of the accuracy of the estimated average μ-value within patients from MR derived μ-maps. Recently, a more optimal initial attenuation image or μ-map estimate in TOF-MLAA reconstruction has been proposed for hybrid PET/MR imaging. The proposed initial μ-map estimate is an image of the object filled with the average μ-value, and the proposed Initial Average Mu-value approach is referred to as IAM-TOF-MLAA. The average μ-value within the object is prior information which can be extracted from MR and patient database. In this work, the accuracy of the average μ-values within patients estimated from simulated MR μ-maps was evaluated using CT μ-maps as references. Clinically acquired human CT μ-maps (10 dedicated brain and 10 whole body scans) were randomly selected and used as the gold-standard. Dixon and non-Dixon MR μ-maps were simulated by replacing bone (and fat) with water as well as assigning a single μ-value for the lungs in the CT μ-maps. The average μ-value was calculated within the head, chest, and abdomen regions from the reference CT, Dixon MR, non-Dixon MR μ-maps for all patients. The error in the average μ-value obtained from the simulated MR μ-maps was evaluated, and correction factors which account for bone, fat, and uniform lungs were derived based on patient data to improve the accuracy of the estimated average μ-value. It was observed that the average μ-value within patients can be accurate within 10% from MR derived μ-map without any correction and within 5% with corrections for bone, fat and uniform lungs. It is also possible to further improve the accuracy of the average μ-value by applying the corrections derived from patients with a more specific range of body mass index and a more sufficient population size.